The Limits of Diversification: Evidence of a Curvilinear Relationship

• In Event: Performance Management, Budgeting Reforms, and Organizational Design

Abstract

Revenue diversification is widely promoted as a best practice for improving nonprofits financial stability, organizational resilience, and overall performance (Carroll and Stater 2009; Mitchell and Calabrese 2023). However, prior research offers mixed and often contradictory evidence regarding its effects, suggesting that the relationship may be more complex than commonly assumed (Hung and Hager 2019; Lu, Lin, and Wang 2019).

This study asks: How does revenue diversification affect nonprofit program outcomes, and are its benefits bounded? Theoretical arguments show that diversification produces both benefits and costs. While spreading revenue sources reduces financial vulnerability (Carroll and Stater 2009), it also introduces administrative complexity that can strain organizational capacity and distract nonprofits from service delivery (Frumkin and Keating 2011). We therefore theorize and test an inverted U-shaped relationship in which diversification improves performance up to a point, after which additional diversification yields negative returns.

We test this hypothesis using a 17-year panel dataset (2008–2024) of U.S. arts and cultural nonprofits drawn from DataArts, comprising over 43,000 organization-year observations. Unlike prior studies that focus primarily on financial indicators, we measure performance using in-person attendance as a proxy for non-financial performance (Kim 2017). Revenue diversification is operationalized using six variants of a Herfindahl-based index capturing different levels of revenue disaggregation (RD3–RD14). We estimate two-way fixed effects models with lagged independent variables to address unobserved heterogeneity and potential endogeneity.

The results provide robust evidence of a curvilinear relationship across all specifications. Diversification initially improves program outcomes, but beyond a threshold, its benefits decline. Importantly, the estimated turning points shift upward as diversification is measured using more disaggregated revenue categories. In the most diversified specification, which distinguishes 14 revenue categories, the relationship becomes positive across its entire feasible range. One possible explanation is that diversification across fundamentally distinct revenue types (such as moving from grants to earned income) imposes higher marginal costs that erode benefits more quickly. In contrast, diversification within similar revenue categories (for example, diversifying among different government levels) allows organizations to leverage existing administrative systems and expertise, thereby achieving higher turning points before diminishing returns set in.

These findings reconcile competing theoretical perspectives from Modern Portfolio Theory, Resource Dependence Theory, and Transaction Cost Economics by showing that each is valid within specific ranges of diversification (Markowitz 1952; Pfeffer and Salancik 1978; Williamson 1981). The research demonstrates that diversification is not universally beneficial but depends critically on organizational capacity to manage administrative complexity (Mitchell and Calabrese 2019; Koerner and Johnston 2022). For practice, results caution against indiscriminate diversification. Nonprofit leaders should match revenue expansion with investments in administrative systems and professional infrastructure. Funders should consider capacity-building grants that help organizations manage complexity rather than imposing requirements that strain limited managerial attention.

Author

• Hala Altamimi, University of Kansas
  Presenting Author